On 09/15/2014 10:31 AM, Daniel P. Berrange wrote:
> On Mon, Sep 15, 2014 at 09:21:45AM -0500, Chris St. Pierre wrote:
>> On Mon, Sep 15, 2014 at 4:34 AM, Daniel P. Berrange <berra...@redhat.com>
>> wrote:
>>> To arbitrarily restrict the user is a bug.
>> QFT.
>> This is why I don't feel like a blueprint should be necessary -- this is a
>> fairly simple changes that fixes what's pretty undeniably a bug. I also
>> don't see much consensus on whether or not I need to go through the
>> interminable blueprint process to get this accepted.
>> So since everyone seems to think that this is at least not a bad idea, and
>> since no one seems to know why it was originally changed, what stands
>> between me and a +2?
> Submit a fix for it, I'll happily +2 it without a blueprint. We're going
> to be adopting a more lenient policy on what needs a blueprint in kilo
> and so I don't think this would need one in that proposal anyway.
> Regards,
> Daniel

What was the behavior in Icehouse wrt to validation here? If we are
talking about the same behavior as Icehouse, I'm fine with a bug fix at
this point in the cycle. If this reverses behavior in Icehouse, I feel
like we should wait until Kilo.


Sean Dague

OpenStack-dev mailing list

Reply via email to